Imagining in Oppressive Contexts, or What’s Wrong with Blackface?*
Abstract
What is objectionable about “blacking up” or other comparable acts of imagining involving unethical attitudes? Can such imaginings be wrong, even if there are no harmful consequences and imaginers are not meant to apply these attitudes beyond the fiction? In this article, we argue that blackface—and imagining in general—can be ethically flawed in virtue of being oppressive, in virtue of either its content or what imaginers do with it, where both depend on how the imagined attitudes interact with the imagining’s context. We explain and demonstrate this using speech act theory alongside a detailed case study of blackface.
I. Introduction
When we tell off-color jokes, entertain sexual fantasies, root for baddies in films, visualize punching obnoxious colleagues, or black up, our imaginations broach morally dicey material. Some of these imaginings clearly produce harmful consequences (or increase their probability).1 In other cases, however, no clear harm ensues. Setting aside consequences, is there anything intrinsically wrong with such imaginings?
Consider two toy examples. When Portsmouth F.C. fans sing, “Hello, hello we are the Portsmouth boys, / And if you are a scummer fan, / Surrender or you’ll die!” they imagine (and prescribe others to imagine) wishing death on Southampton F.C. fans (“scummers”) who refuse to support Portsmouth. Participating fully in the song involves imaginatively adopting the attitude that loyal Southampton fans deserve death. Sincerely adopting such an attitude would be unethical. But condemning someone who sings the song just “for fun” seems puritanical.
But now, compare this to the Sigma Alpha Epsilon (SAE) fraternity song that made headlines after being caught on video in 2015: “You can hang him from a tree, / But he can never sign with me, / There will never be a n----- in SAE!” Singing this song was widely condemned—rightly so, it seems, even if the song’s attitudes were only imaginatively adopted, even if we unrealistically stipulate that the singing caused no harmful consequences, and not merely because the song includes a racial slur.
What explains the difference when both examples involve imaginatively adopting unethical attitudes? Answering this question furnishes a reply to our titular query. We argue that an imagining is ethically flawed when it oppresses in virtue of its content or what imaginers do with it. Both kinds of flaw depend on how the imagined attitudes interact with the imagining’s context. In this respect, imaginings are like speech, an analogy we exploit throughout. Put precisely in terms yet to be fully explained, imaginings whose deployed attitudes realize a “controlling image” (a representation ideologically supporting an oppressive system), normalize oppression, or license oppressive behavior are ethically flawed. This holds whether or not such imaginings thereby endorse these attitudes or cause subsequent harm.
We illustrate this general account with a detailed philosophical case study of blackface, a practice whose moral contours warrant their own article. While criticism of blackface dates back to at least the nineteenth century, its prevalence and persistence continue to generate controversy and confusion. Consider these events from 2019 alone. Canadian prime minister Justin Trudeau was found to have once sported “blackface” and “brownface” (i.e., skin-darkening makeup intended to make him look Black and Middle Eastern, respectively). Former Virginia governor Ralph Northam, former attorney general of Virginia Mark Herring (both Democrats), and Alabama governor Kay Ivey (a Republican) all admitted to having worn blackface in their youth for a dance contest, rap performance, and church skit, respectively. Celebrity Kim Kardashian was criticized for being photographed in lighting that made her appear Black. Fashion house Gucci released and then recalled a black balaclava sweater with red lips around the mouth opening. Renowned ballerina Misty Copeland criticized the Bolshoi Theatre for its continued use of blackface. Finally, the Netherlands revisited the national debate over “Zwarte Piet” (Black Pete), a traditionally blacked-up character portrayed in the annual Sinterklaas festival. Clearly, it is worth articulating what, exactly, is wrong with blackface.2
We begin in Section II by situating our view dialectically within a recent debate in aesthetics. In Section III we analogize speech and imagining. In Section IV we explain how sociohistorical context can make imaginings oppressive, and in Section V we explain how relevantly critical (e.g., parodic and educational) imaginings may avoid this flaw.3 We close in Section VI by rebutting some potential objections. Throughout (as mentioned), we apply our theory to the interpretatively difficult, concrete case of blackface to bring out our proposal’s nuances.
II. The Ethics of Imagining
To understand the ethics of imagining, it is fruitful to look first at the ethics of artworks. Aestheticians distinguish two ways artworks can be unethical: intrinsically and extrinsically. Representational artworks ask appreciators to adopt certain attitudes—they “prescribe” these attitudes, as is said—in order to fully appreciate the work.4 A fantasy novel, for instance, might prescribe (imagined) beliefs that dragons exist, (imagined) fears of them, and so on.5 Artworks can possess intrinsic ethical properties owing to how they “manifest,” “express,” “promote,” or “call upon” certain moral attitudes via these prescriptions.6 For instance, The Birth of a Nation,7 a film that spurred African American communities to organize nationwide protests, is intrinsically ethically flawed because it prescribes racist attitudes. By contrast, artworks exhibit extrinsic ethical flaws when the fault lies in the causal consequences of appreciating them (e.g., the copycat crimes inspired by A Clockwork Orange8) or their causal etiology (e.g., the Louvre Abu Dhabi’s exploited construction workers9). Aestheticians broadly agree that ethical criticisms of artworks as such should only target their intrinsic features; features such as causal etiology generally relate to works too adventitiously to ground evaluations of them qua art.10
This framework naturally extends to imaginings, which include but go beyond artworks. As we use the term, “imagining” refers to all imaginative content and all imaginative acts. Imaginative content includes daydreams, fantasies, and works of fiction, among other things. Imaginative acts generally realize this content inside the head (e.g., daydreaming, fantasizing, reading a novel), outside it (e.g., writing a script, painting a portrait), or both (e.g., playing cops and robbers).11 For brevity, we will refer to an imagining’s centrally involving or prescribing attitudes as “deployment”; privately fantasizing that one is weightless, or a novella in which this is so, for instance, deploys the (imagined) belief “I am weightless.” Just as artworks can manifest an ethically criticizable (or laudable) character by deploying such attitudes, then, so can imaginings.
It is important to our argument that some intrinsic features of imaginings depend on contextual facts. One might protest that it makes no sense to speak of a context-dependent intrinsic feature. Are a thing’s intrinsic features not precisely those that depend on nothing beyond that thing? Let us distinguish two senses of ‘intrinsic’. Call an imagining’s feature “strongly intrinsic” if it is intrinsic in the robust metaphysical sense that its existence depends on nothing besides the imagining, that is, it inheres in the imagining nonrelationally.12 Many features that aestheticians discussing artworks have described as simply “intrinsic,” such as gracefulness, brevity, or goriness, may be strongly intrinsic. A “weakly intrinsic” feature, by contrast, also inheres in the imagining but arises from interactions between the imagining’s other (strongly or weakly) intrinsic features and other entities or states of affairs. Many paradigmatic intrinsic features of imaginings will count as weakly intrinsic on this characterization. That Artemisia Gentileschi’s Judith Slaying Holofernes is executed in oils or represents three people,13 for instance, might be a strongly intrinsic feature of the imagining.14 The imagining’s being innovative or continuous with a baroque tradition, however, is certainly a weakly intrinsic feature, partly grounded in works that preceded its painting. Since imaginings are the products of human meaning making, context inevitably conditions what they are and mean. The distinction we draw between (weakly) intrinsic and extrinsic features, then, following a tradition in aesthetics, is not the metaphysician’s. It is a distinction between what belongs to a thing (e.g., an artwork, an imagining) and what is merely intimately connected with it. The thought is simply that whether a painting, say, celebrates vice is a question about the painting, whereas whether appreciating it causes harm is a question about something else.
Numerous scholars have argued or assumed that imaginings, like artworks, exhibit intrinsic ethical flaws when they deploy imagined unethical attitudes. Call this view “imaginative strictness.”15 Others, however, have recently criticized this position. They argue, roughly, that any imaginings not also endorsing these attitudes, by recommending their serious adoption in the actual world, fall outside moral evaluation’s scope.16 Call this view “imaginative laxity.”17
To appreciate this debate, we must distinguish two senses of ‘prescription’, which the literature uses ambiguously, or, correspondingly, two senses of ‘deploy’.18 First, imaginings can deploy attitudes assertorically, as it were, as when someone vividly imagines losing a limb to learn how a recent amputee feels. Imaginings deploying attitudes in this way are to that extent constrained by the imaginer’s beliefs and, correspondingly, present the imagined attitude as suitable for “export” into the actual world.19 We sometimes engage in imaginings of this kind to better understand the actual world, as when reasoning counterfactually or visualizing the past. We might, for example, imagine piles of bodies during Indonesia’s invasion of East Timor, to better grasp the US government’s decision to green-light it and the ensuing genocide.20 Second, imaginings can deploy attitudes in a way that requires “prescind[ing] from any alethic commitments” to what is imagined.21 Hence, to fully participate in a comedy skit in which US secretary of state Henry Kissinger weepily implores Indonesian president Suharto to “think of the children” and stop the invasion, we must suspend some of our true beliefs to play along. We call the former sense “endorsement”; the latter, “fictive deployment.”22
Using our terms, then, the dialectic is this. On the strict view, imaginings exhibit intrinsic ethical flaws whenever fictively deploying or endorsing morally objectionable attitudes. Meanwhile, the lax claim that imaginings exhibit these flaws only when endorsing such attitudes. We tread a middle path. We agree with the lax that imaginings may be intrinsically unethical in virtue of the attitudes they endorse. However, we disagree that they must endorse an attitude to suffer such a flaw. Performing the SAE song is morally problematic in our present sociohistorical context, for instance, even if the song does not endorse the racist attitudes fictively deployed. We agree with the strict, meanwhile, that fictively deploying unethical attitudes can generate intrinsic ethical flaws. However, we disagree that it must. There is nothing intrinsically objectionable about singing the Portsmouth football song, for instance. We argue that imaginings like the SAE song, and unlike the Portsmouth football song, exhibit a kind of intrinsic ethical flaw in virtue of doing one of three things: fictively deploying attitudes that realize oppression, normalize oppression, or license oppressive behavior.
Beyond this scholarly debate, our arguments have a broader target: the proto-laxity popular among the general public. Many are convinced that imaginings trading in problematic attitudes are morally acceptable provided that they are undertaken in jest or without derogatory intent. Consider the phenomenon of “hipster racism.”23 Here, social progressives satirically make jokes trading on racial stereotypes, appropriate non-White culture,24 briefly visit predominantly non-White spaces (the “ghetto”) for personal amusement, and use racial slurs purportedly to challenge social norms. In doing so, they assume that, since they obviously do not endorse racist attitudes, these behaviors are merely ironic or edgy, or manifest cross-racial solidarity. Or consider the University of Oregon professor who dressed in blackface at a Halloween party for her students to celebrate an antiracist book, Black Man in a White Coat.25 Most such examples, where they involve the imagination, do not endorse racist attitudes. Nevertheless, they often fictively deploy them. Both cases have been sharply criticized as morally problematic and distressing to people of color.26 On the lax account, there is nothing objectionable about merely fictively deployed racist attitudes; so, this distress is unwarranted as a response to the imagining qua imagining—it appears oversensitive, sanctimonious, or like political correctness run amok. Yet our account vindicates at least some of this distress by identifying a moral problem warranting it.
III. Imaginings and Speech
Our starting point is J. L. Austin’s distinction between an utterance’s three dimensions: locutionary, illocutionary, and perlocutionary.27 Austin’s aim is to peel away both an utterance’s content (locution) and its causal upshot (perlocution), to reveal the act the utterance constitutes (the illocution). Consider his example of A telling B, “Shoot C!” The locutionary act is communicating that B is to shoot C; the perlocutionary act, if A succeeds, is B’s shooting C (subsequent perlocutionary effects include C’s being shot, dying, etc.); the illocutionary act is that which is constituted or enacted by the utterance: in saying “Shoot C!” A urges or orders B to shoot. Importantly, the illocutionary force of A’s speech act may apply whether or not A succeeds in causing B to shoot.28
Feminist and critical race philosophers have used speech act theory to illuminate oppressive speech. Catharine Mackinnon, Rae Langton, and Jennifer Hornsby, for instance, have used it to turn the “free speech” defense of pornography on its head: if pornography is speech, then it can have the illocutionary force of subordinating and silencing women, furnishing grounds for ethical (if not legal) condemnation.29 Luvell Anderson, Sally Haslanger, and Rae Langton have also identified various racially oppressive speech acts: propaganda-like hate speech that incites or promotes racial violence and discrimination, assault-like hate speech that persecutes and attacks members of certain racialized groups, and speech that subordinates or discriminates in virtue of some institutional authority.30
Speech act theorists have long recognized the importance of context. Often, without grasping the contexts in which speech takes place, one cannot know what content an utterance has. What saying “they are here” literally conveys, for example, will depend at least on which group is salient and the speaker’s location. Additionally, and more generally, which action an utterance constitutes is also context dependent. As Austin and others note, even where one fully grasps an utterance’s literal meaning (as a locutionary act), one may not know what is being done with it (as an illocutionary act).31 The same claim “Your coat’s on the floor” could be an (illocutionary) act of informing someone that she has dropped her coat, of indirectly commanding her to pick it up, or of replying to a question about the coat’s whereabouts. Since context can determine an utterance’s locutionary content and illocutionary force, it can also determine whether and how an utterance warrants ethical opprobrium.
We argue that imaginings, again understood as a broad category of representation, are importantly analogous to speech. We neither claim nor presuppose that imaginings are speech, as some do.32 We maintain only that imaginings resemble speech in important ways. First, many imaginings deploy propositional content with which one can perform various types of actions and cause downstream effects. If, beholding her right hand, a tourist imagines that it is Michigan, this “locutionary” act has propositional content: it makes the statememt “I am looking at Michigan” true in the make-believe game she thereby plays. And if she examines the crook of her thumb and thinks, “Ann Arbor is here,” she not only imagines something; she thereby also performs the “illocutionary” act of assenting to something about Ann Arbor’s actual whereabouts. This, in turn, could have “perlocutionary effects” on how she navigates around Michigan. Similarly, novels, sculptures, daydreams, ad pitches, and so on, make certain propositions true in a “fictional world,” may assert those (or other) propositions,33 and may causally affect appreciators. George Orwell’s Animal Farm,34 for instance, makes it fictional that a farm animal revolution is corrupted, thereby asserts that noble ideals can be harnessed for oppressive ends, and has impacted generations of readers. Second, which attitudes an imagining deploys and, to a greater extent, what sort of act it constitutively embodies are generally intrinsic features of the imagining partly determined by context. Animal Farm’s political backdrop, for instance, helps make it political criticism rather than mere fairy tale.
Thus, in addition to representing content (as locutionary acts), causing effects (as perlocutionary acts), and exhibiting the kind of context sensitivity described, imaginings may also constitute certain kinds of acts analogous to illocutionary ones. Like utterings, some imaginings are doings.35
IV. Constituting Oppression: Realizing, Normalizing, and Licensing
Following several authors, we take oppression to consist, roughly, in a system of interlocking disadvantages applying to members of a socially defined group qua members of that group, where this disadvantage spans multiple domains of social life.36
We can now fully explicate this part of our thesis: imaginings oppress insofar as they fictively deploy attitudes in virtue of which they (a) realize oppression, (b) normalize oppression, or (c) license oppressive behavior.37 Realizing oppression concerns the content of an imagining and corresponds, on our analogy, to the locutionary dimension of speech; imaginings realize oppression by partially constituting an oppressive system. Imaginings normalize oppression by rendering the unjust treatment of certain social groups (seemingly) justified, or so natural and “given” that it needs no justification. They license oppressive behavior when, in fictively deploying the kinds of attitudes that normalize oppression, they activate oppressive norms in particular microcontexts. Normalizing and licensing concern what is being done with the imagining and to that extent represent the imagining’s “illocutionary” dimension.38
It is worth stressing two points here. First, realizing, normalizing, and licensing unjust treatment are ways of constituting rather than causing oppression.39 To be clear, we are highly sympathetic to the view that the kinds of imaginings we discuss—not just blackface but pornography, stereotyped characters, derisive jokes, and so on—can be criticized in terms of indirect harm or risk thereof. But here we argue that these imaginings are ethically flawed not extrinsically but intrinsically. This claim can hold even when harmful consequences are contested, difficult to detect, or simply nonexistent.
Second, and relatedly, whether an imagining’s realizing or normalizing oppression or its licensing oppressive behavior itself constitutes a harm depends on how broadly one understands ‘harm’.40 On one influential liberal account, for instance, harms are setbacks to interests induced by a wrong, as when one makes someone poorer by stealing their wallet.41 But this itself admits of more or less broad readings depending on how one understands ‘setbacks’ and ‘interests’. Nor is the question settled if, as seems likely, oppressive systems are ipso facto harmful. For, even so, it does not follow that all of its constitutive elements are. Assuming otherwise commits the fallacy of composition. The oppression of Black people in the Jim Crow South, for instance, was harmful. And hanging a “Whites Only” sign in a shop window in that context would be oppressive. But whether the sign’s hanging as such is a harm is a further question that swings independently of the harmfulness of both the oppressive system of which it forms a part and the sign’s harmful effects. After all, these effects may be none if the sign is hung incompetently and never even seen. What we show is that imaginings suffer a pro tanto moral defect. Whether they do this in virtue of being harms is a further question on which we take no position. Again, our aim is to demonstrate that imaginings like blackface are ethically flawed even if we grant, for the sake of argument, that no harm has been caused by them.
We proceed stepwise. In Section IV.A, we show, first, how blackface’s history helps determine which attitudes its use fictively deploys. These realize oppression by partly constituting an oppressive system. Next, we argue that fictively deploying such attitudes, if only in certain contexts, amounts to using them in oppressive ways, even where no harm ensues: normalizing oppression (Sec. IV.B) and licensing it (Sect. IV.C). These represent distinct mechanisms by which sociohistorical context alters the ethical status of imaginings.
A. Realizing Oppression: Controlling Images
Sociohistorical context can affect imaginings’ ethical status “locutionarily” by determining which attitudes they fictively deploy, that is, their content. In the case of contemporary blackface, we argue, the attendant imaginings’ content is largely determined by blackface’s historical uses and the meanings attached to the practice as a result.
The extension of “blackface” unquestionably includes the classical minstrel performances that emerged in the United States two centuries ago. Here, performers made up their faces with burned cork, wide red or white lips, and so forth. It has also come to refer, as our opening examples indicate, to virtually all cases in which people of whatever complexion darken their skin to impersonate Black people. This includes practices of blacking up that share, as it were, a phylogenetic branch with minstrelsy, such as the “Ghanaian Concert Party,”42 which take place in societies relatively distant from the originating US cultural mainstream. But we take the practice to be broader. At least provisionally, it includes stereotyped vocal impersonations, such as Robert Downey Sr.’s voice dubbing in Putney Swope, as well as buffoonish representations of Black stereotypes without literal skin darkening, such as the later Amos ’n’ Andy television show, to take just two examples.43 These examples, anchored by the paradigm case of minstrelsy, suffice for our argument.
Blackface minstrel shows, from their beginnings in the early nineteenth century or even late eighteenth century in the United States,44 were overwhelmingly used to present a fantastical and mostly derogatory conception of Blackness. Though plucked from the White imagination, this conception was nevertheless presented as faithfully reproducing the lives and traditions of Black people.45 It was propagated through Jim Crow, the Black Dandy, Zip Coon, Lucy Long, Mammy, Jezebel, Tambo, Bones, Uncle Tom, Pickaninnies, and other stock characters. Thus, the “locutionary” content of the imaginings in blackface minstrelsy was clear: Black people are some combination of lazy, obsequious, ignorant, pretentious, sexually promiscuous, voracious, happy-go-lucky, and content with their own oppression, among other things. We will refer to this complex of attitudes as “anti-Black.”
Blackface minstrelsy went on to exert tremendous influence over American and global culture and entertainment, across vaudeville, film, radio, television, music, and literature.46 It did so in tandem with or by infusing anti-Black representations in these other genres and media. This is exemplified by plays such as Blanche of Brandywine,47 “bobalition” broadsides, cartoons, and films—including, of course, The Birth of a Nation. Mickey Mouse, figurehead for one of history’s most influential global entertainment enterprises, embodies this subterranean influence. Besides starring among “savages” and other racist tropes in the early cartoons, Mickey is himself a minstrel, at least a vestigial one. He is a mischievous half-agent, half-object, fated to sing and dance in a world of violence and buffoonery with his wide-lipped, gaping smile and telltale white gloves.48
It might be objected at this point that the precise meanings of blackface, including the attitudes it fictively deploys, are complicated and continually contested. Minstrelsy was put to many different uses, including some transgressive, resistant, or even emancipatory ones. Interpretive paradigms have thus shifted throughout decades of scholarly work, much as the practice itself has changed over time.49 Moreover, Black people themselves have been not only performers but also, at times, avid enjoyers of the genre. Ethnically Jewish performer Al Jolson’s blacked-up starring role in The Jazz Singer,50 for instance, endeared him to a large African American audience and earned him praise in the African American–owned press.51 Given this complex pentimento of meanings, one might question whether minstrelsy or blackface in general essentially involves the deployment of anti-Black attitudes—surely not, if even Black audiences warmed to minstrels like Jolson? Perhaps, the skeptic might continue, contemporary abhorrence to blackface reflects a skewed interpretation of the practice’s history and meaning, or, worse still, the “moral grandstanding” of White liberals.52
While not without its merits, this line of thought misses the fact that, for all the historical complexity, the main artery pulsing through blackface’s history is profoundly anti-Black. This is conceded even by historians with more sympathetic readings of minstrelsy. For even when blackface minstrelsy was used to espouse abolitionist sentiments, for instance, these were not devoid of anti-Blackness. As Frederick Douglass noted in an 1856 editorial, “opposing slavery and hating its victims has come to be a very common form of Abolitionism.”53 And despite minstrelsy’s rough similarities to some authentic African American cultural traditions, it nevertheless remained—as Ralph Ellison shrewdly observes—a White artistic form, as evidenced by the requirement that Black performers themselves black up.54 Hence, African American admiration for Al Jolson is better explained by his public image. Jolson was seen as someone who enjoyed friendly race relations, worked in a theatrical tradition enjoying significant African American participation, and tried to materially improve the lives of the Black artists with whom he worked.55 More generally, many people enjoy entertainment that, on analysis, derogates them. In the case of blackface, this observation was made as early as 1841 by African American journalist Samuel Cornish. Cornish’s lament at the number of African Americans enjoying performances that would “heap ridicule and a burlesque upon them in their very presence, and upon their whole class” has echoed throughout the Black intellectual tradition up until the present.56
A crucial fact about blackface is that its default meaning changed drastically during the US civil rights movement. Though critiques of minstrelsy are traceable to the mid-nineteenth century, the denunciation of blackface by prominent figures such as Ralph Ellison and LeRoi Jones during the 1950s and 1960s transformed the public understanding.57 Just as Confederate statues were transformed into unambiguous symbols of White supremacy once erected in defiance of civil rights advancements, so too did blacking up become definitively marked (barring critical exceptions) as anti-Black expression. So, while blackface minstrelsy may have served resistant and transgressive ends in the past, this has become significantly more difficult following the historic change in meaning.
This history has profound implications for understanding why blackface almost always fictively deploys anti-Black attitudes (again, see Sec. V). Much as word meaning depends on histories of use and wider communicative practices, so too do some imaginings’ history of use determine their meaning. One cannot simply use ‘dog’ to refer to cats, for instance. Likewise, one cannot uncritically engage in certain imaginings without fictively deploying their standard attitudes. This is because imaginings can and often do fictively deploy attitudes even where imaginers lack any knowledge or intent concerning them. This is clearly true of the University of Oregon professor. In blacking up her face, she evidently intended to fictively deploy the belief that she was a Black man and no derogatory attitudes. Nevertheless, the historical context in which this act took place is not only one where such imaginings had fictively deployed and endorsed precisely such derogatory attitudes; it is also one where, as we explain shortly, these attitudes still perform their hegemonic function. Accordingly, the professor could scarcely avoid her imagining deploying these anti-Black attitudes.
We can now demonstrate the first way that sociohistorical context makes imaginings ethically criticizable, which brings us into the realm of social ontology. The point is this: in a different context, imaginings that fictively deploy anti-Black attitudes need not be oppressive. But they are in ours, because of the constitutive role this imaginative content plays as purported justification for the present system of anti-Black oppression. More generally, we propose that imaginings realize oppression whenever they instantiate what sociologist Patricia Hill Collins calls a “controlling image.”58 According to Collins, controlling images are stereotypes, symbols, and other portrayals of oppressed social groups (“ghetto chick,” “dangerous thug,” etc.) “designed to make racism, sexism, poverty, and other forms of social injustice appear to be natural, normal, and inevitable parts of everyday life.”59 Put simply, controlling images are components of pernicious ideologies that function to maintain unjust social orders.60 When imaginings instantiate controlling images, they often causally perpetuate oppression by leading people to behave oppressively. Yet, even absent these effects, they still partially constitute, or realize, systems of oppression.
This is evident from the metaphysics of social structures and systems. Social theorists such as William Sewell and Sally Haslanger distinguish their “material” from their nonphysical “semiotic” (“symbolic,” “ideological,” etc.) constituents. The former refer to physical objects or situatedness in the physical world; the latter are sometimes described as “virtual.”61 Controlling images are elements of this virtual stuff, which is indispensable for holding the system together. Eduardo Bonilla-Silva, in his account of the social structure of racism, puts it bluntly: “The prejudice of individuals is not, and can never be, the basis for maintaining racial inequality; without an ideology to justify and enable racial projects, racial domination would not be possible at all.”62
Collins offers a particularly rich theory for understanding anti-Black oppression. She identifies a system of four distinct but interrelated domains of power across which certain social groups are socially disadvantaged. The structural domain concerns the laws, institutional policies, and practices in employment, government, education, law, business, and housing that distribute social resources inequitably. The disciplinary domain concerns the systems of bureaucracy and surveillance with which the structural domain’s operations are managed. The interpersonal domain concerns everyday practices by which people (mis)treat one another—including microaggressions as well as overt abuse—and the individual attitudes attending them. The systematicity of unjust treatment across all these domains depends on and is unified by an ideological glue of “commonsense ideas”: beliefs, representations, stereotypes, and so on. These constitute the fourth, hegemonic domain.63 Only when these domains interconnect and draw from the same ideological fount to privilege members of some groups and disadvantage others does oppression obtain.64 Any token disadvantage from one domain disconnected from processes of disadvantage in the others may be morally bad or not. But it does not constitute oppression. In short, controlling images are partially constitutive of oppression in virtue of fulfilling this hegemonic function.
Clearly, then, whether something is a controlling image or not necessarily depends on its context. Unlike mere wrongdoing, oppression does not consist in isolated acts of harm or domination. As Marilyn Frye famously argues, oppression is like a birdcage. A bird confronting a single wire circumvents it easily; when those wires join together to form a cage, however, the bird is trapped. Oppression is social disadvantage unjustly and systematically patterned across virtually all aspects of society.65
So much for the general account, which applies to all imaginings (whether visual, musical, dramatic, literary, etc.) that contain controlling images. What about blackface specifically? Blackface is an imaginative practice that arose and persists against a background of anti-Black oppression. People racialized as Black in the United States (as elsewhere) are oppressed in virtue of being disproportionately excluded from employment, housing, health, education, and other social institutions (structural); being subject to surveillance, marginalization, and unevenly applied standards even when included in organizations (disciplinary); and experiencing prejudice in everyday interactions with others (interpersonal). Systematic Black disadvantage across all these domains depends on our ability to delineate a socially constructed group of “Black people” in the first place. Courtesy of the hegemonic domain, we understand this group to largely comprise “ghetto chicks,” “dangerous thugs,” and so on. To engage in an imagining that fictively deploys these controlling images or other anti-Black attitudes, then, is to perform an act that realizes an oppressive system by partially constituting it.
B. Normalizing Oppression
In what way are these images controlling? Answering this requires us to examine imaginings’ “illocutionary” dimensions. Here the emphasis is not on oppressive systems’ socio-ontological constituents but on the dynamics via which agents maintain them. As with speech acts, many acts can in principle be performed with imaginings like blackface. Indeed, Bonilla-Silva identifies five distinct functions of racial ideology.66 We will discuss two: normalizing and licensing (in this subsection and the next one, respectively).
Because controlling images function to make oppressive systems appear normal, natural, and needing no justification, they can be used to normalize oppression. That is, they dispose participants to acquiesce to current social conditions, often implicitly. But while this is a kind of effect of the imagining, it is not a straightforwardly causal effect.67
To illustrate, consider Mary Kate McGowan’s discussion of how we sometimes enact rather than cause certain facts about the world.68 For instance, simply by sitting outside the Swedish parliament, Greta Thunberg made the claim “Greta Thunberg sat outside the Swedish parliament” true. By uttering the words “Entire ecosystems are collapsing!” Thunberg made “Greta Thunberg said ‘Entire ecosystems are collapsing!’” true. Thunberg did not cause these facts to obtain in the way anthropogenic climate change (partially) did; rather, she enacted them. As McGowan puts it, “Simply by being and doing things, we thereby routinely affect what is true of the world and thereby enact these truths about it.”69
We argue that this is what happens when a person engages with imaginative content containing controlling images: simply by doing so, she makes such portrayals of social groups manifest in actual social life. In other words, by instantiating without challenging a controlling image, she contributes to making it a fact that a social group is (socially) viewed in that way. Moreover, she does so regardless of whether she endorses the portrayal. This enactment of the fact that people routinely view a social group in ways that would (if true) rationalize their unjust treatment is what we call normalization. Because these images are normalized and pervasively “out there” in the world, the differential treatment suffered by oppressed groups appears normal, natural, and either justifiable or in no need of justification at all. And each time a controlling image is instantiated, the more normalized it becomes. In this way, controlling images are akin to a “desire path” impressed across a meadow. Each rambler that treads the path further establishes it. With time, it becomes normal and justified to follow the path and, by the same token, aberrant to tramp through the surrounding grass.
Unlike many other illocutionary acts, normalizing requires no special authority (see also Sec. IV.C). Anyone can directly enact such social facts. As social actors, we shape social reality simply by being and doing things. But by enacting such facts about social groups in a context where others similarly do so, one also makes it the case that unjust practices appear normal. For instance, by imagining a “ghetto thug” in our world, one thereby contributes to making Black men be seen as dangerous and violent. This makes it the case that disproportionate police brutality against Black men appears normal. This is so even when there is no immediate harmful consequence or readily traceable causal process by which one’s imagining subsequently produces harm.
It is in this sense that such images function as instruments of “control” over dominated groups: by impeding their unjust treatment even coming into question. It bears noting that “controlling images” include more than just negative stereotypes; the traits they ascribe to a group may not be negative in themselves, though they serve to rationalize its unjust differential treatment all the same. For instance, recalling the complex of anti-Black traits fictively deployed in blackface, there is nothing bad about being happy-go-lucky. But this becomes weaponized in the context of racial oppression.70 As historian Blair Kelley, discussing relatively early minstrel performances, summarizes, “These performances were object lessons about the harmlessness of southern slavery. By encouraging audiences to laugh, they showed bondage as an appropriate answer for the lazy, ignorant slave. Why worry about the abolition of slavery when black life looked so fun, silly, and carefree?”71 As leading minstrel performer Thomas “Daddy” Rice opined to an audience in 1837, his Jim Crow character “effectually proved that negroes [sic] are essentially an inferior species of the human family” who “ought to remain slaves.”72 In 1865, a South Carolinian slaver publicly admitted in a letter that he had been “laboring under a delusion” that “these people were content, happy, and attached to their masters.” He was disabused of this delusion by the mass exodus of enslaved people from plantations at the conclusion of the US civil war.73 To be sure, this case exemplifies the downstream “perlocutionary” effects caused by blackface minstrelsy. But it also highlights the noncausal normalization of oppression via such imaginings; it is hard to see how the South Carolinian slaver could have believed what he did unless he was in the grip of controlling images. Our view, then, is that blackface is intrinsically ethically flawed, whether or not a given instance produces harmful consequences, because it instantiates (without negating) controlling images that normalize anti-Black oppression.
One implication of our view is that blacking up for imaginative purposes before the established tradition of blackface arose would not be criticizable for the reasons documented here, namely, of realizing oppression or constituting an oppressive act (though it might be for others). An actor blacking up in an early seventeenth-century production of Shakespeare’s Othello,74 before modern racial ideology had fully crystallized, for instance, would not fictively deploy the same anti-Black attitudes as an actor doing so today. However, he may fictively deploy attitudes reflecting Shakespearean-era prejudices, of course.
Post–Jim Crow, however, blacking up has assumed a different and highly objectionable social meaning that foregrounds its entanglement with anti-Black systems of slavery and colonialism (see Sec. IV.A). The nascent recognition of this helps explain the growing trend of casting Black actors as Othello or eschewing blackface.75 It also explains reformations of traditions in which it remains unsettled whether painting the face black represents race at all, such as in morris dancing.76 For the same reasons, recent years have witnessed increasing protest against blacking up as Zwarte Piet in the Dutch Sinterklaas festivities, as well as Balthazar in the Spanish-origin Reyes Magos celebrations.77 Such traditions may not have been problematic when they began (at least not for precisely the same reasons). Yet the context of our world, where this meaning remains significant, makes the imaginings they deploy objectionable. Current efforts to reform such traditions reflect a growing recognition of imaginings’ importance in normalizing anti-Black oppression. They also reflect increased globalization and the merging of once-distinct sociohistorical contexts.
To sum up, fictively deploying certain derogatory attitudes in relevant macrocontexts is ethically flawed whenever those attitudes constitute controlling images. This holds even when the imaginings in question neither endorse these attitudes nor cause harm. To engage in such imaginings is to mobilize these images’ hegemonic function of normalizing existing systems of oppression. However, this flaw is not exhibited by all imaginings that deploy morally objectionable content. Insofar as the attitudes do not hook up with existing forms of structural, disciplinary, and interpersonal domination, no such flaw obtains.
C. Licensing Oppression
A third way that sociohistorical context conditions imaginings’ ethical status is by making certain imaginings perform the “illocutionary” function of licensing oppressive behavior.
Here, we again appeal to McGowan’s body of work explaining how ordinary speech can oppress.78 For McGowan, utterances are acts of oppression whenever they, as Robert Simpson puts it, “bring the latent force of [an oppressive] system to bear in the local context” in which they occur.79 Social actors do this in virtue of their ubiquitous power to enact not just facts but also norms. Consider an example: when a coach declares, “No phones during practice!” she thereby makes it the case not only that she uttered those words but also that it is now inappropriate for athletes to use their phones. Here, the coach has special authority to set rules on the playing field; her declaration enacts the norm against phones. Players could believe otherwise, but their expressing this does not cancel the norm, no matter how devastatingly phrased the disgruntled tweet.
However, McGowan demonstrates that in ordinary conversation speakers need no special authority to enact oppressive norms; their power obtains simply in virtue of how conversations work. In conversation, each contribution a speaker makes alters her interlocutor’s ways of permissibly continuing the conversation. When someone asks a question, for instance, this demands an answer; responding with a non sequitur on an unrelated subject is no longer an appropriate “move” for that particular conversation.80 This enactment of norms governing only the specific microcontexts in which they occur is what we call licensing. So, when someone asks, “Do you support Portsmouth or Southampton?” she thereby licenses a relatively narrow range of utterances which now count as legitimate next moves within this particular conversation, for example, “Portsmouth,” “Southampton,” “neither,” “I don’t follow football,” and so on. It would be inappropriate, by contrast, to respond by listing one’s favorite Nicolas Cage films, declaring war on Kyrgyzstan, reciting the lyrics to Meat Loaf’s Bat Out of Hell, and so on.
McGowan argues that gender, racial, and other oppressions resemble conversations in that, in a sense, which actions are socially appropriate is set by prevailing social norms of unjust treatment. Racial oppression, for instance, is a complex of social practices wherein treating people of color as inferior counts as an appropriate move in the “game,” while treating them as equals is inappropriate. White supporters of the US civil rights movement, for instance, often suffered job loss, physical violence, and other reprisals for violating these “rules.” Specific instances of sexist and racist speech, then, license subsequent unjust treatment when they bring such norms to bear on the microcontext of a particular social interaction. For example, a speaker telling colleagues a sexist joke thereby alters the norms governing that social interaction. He thereby makes it conversationally appropriate for them to demean women by laughing at the joke, swapping another for it, and so on.81 Importantly, this is oppressive whether his colleagues actually respond in kind or not. The joke constitutes oppression merely by enacting norms that render sexist treatment socially appropriate in that interaction. This is independent of whether the joke subsequently causes actual unjust treatment.82
Mutatis mutandis, the same is true, we claim, of imaginings.83 By bringing the force of an oppressive system to bear on a particular microcontext, imaginings license oppressive behavior. This applies to imaginings of any kind, be they fictional stories, games, jokes, visual representations, and so on, or uses of these. They do this by enacting unjust norms of treatment in the specific microcontexts in which they occur. This, then, is a further way that imaginings can be used to perform oppressive acts—though, again, only within wider macrocontexts with relevantly oppressive background conditions. To show this, we again return to our central test case of blackface.
Before applying McGowan’s theoretical machinery to blackface, we must first acknowledge a potential complication. Blackface minstrelsy has been used to perform a variety of (illocutionary) acts, not all of them morally reprehensible. Despite its racially derogatory meanings, the practice has occasionally been used by members of the Black community to advance their own ends, for example, by Black entertainers who themselves performed in minstrel shows.84 “Playing Black” often granted—indeed still grants—African American artists access to artistic industries, as Spike Lee’s Bamboozled fictionalizes so forcefully.85 Many have also noted the practice’s capacity for having allowed Black artists to hone and demonstrate their craft. Artists such as Bert Williams, for instance, exploited the artform for real creativity, moving beyond, or even satirizing, some of its racist tropes.86 Blackface has also functioned as a protective screen permitting the oppressed to avoid insult and violence, even if also distorting the expectations of Whites who would inflict such mistreatment.87 The African American entertainer Tom Fletcher and his minstrel troupe, for instance, would stay in character after performances in Southern towns, parading from the theater to the train station to the minstrel standard “Dixie.” This averted the anger of Whites intolerant of Blacks acting “out of character.”88
Blackface was also sometimes used by Black and White performers alike to mock other more reasonable targets—albeit generally via the mockery of an imagined Blackness. These included the floral oratory, sartorial pretensions, and general extravagance of the ruling classes.89 Thus, mock-Blackness sometimes provided a ridiculous cover for speaking truth to power.90 In the early nineteenth century, minstrel show audiences were mostly poor and working class. The affirmation of “low” culture, alongside abolitionist sentiments and the transgression of conservative sexual and gender norms, countered the efforts of elites to discipline them into a compliant industrial workforce.91
In addition to these worthier ends, much of White audiences’ attraction to minstrelsy was plausibly rooted in a genuine admiration for and identification with an emergent African American culture, however misrepresented the culture and problematic the admiration.92 The complicated uses of blackface partly explain why many who enjoyed these performances were themselves Black. Of course, the fact that Black and White audiences laughed at some of the same shows does not mean that they did so for the same reasons.93 Nevertheless, the historical record suggests that in some instances blackface performances fostered cross-racial solidarity.
While this historical complexity is important and often forgotten, we nonetheless contend, as in Section IV.A, that the solidification of oppressive meanings following the civil rights movement has made uses of blackface today nearly always oppressive. Consider the 2010 “Compton Cookout” party hosted by a University of California fraternity. Female partygoers were asked to imitate so-called “ghetto chicks,” described as wearing “cheap weave, usually in bad colors”; having “a very limited vocabulary”; and “making other angry noises, grunts, and faces.”94 In short, the Compton Cookout party fictively deploys contemptuous attitudes toward African Americans according to which they are unintelligent, uncultured, aggressive, vulgar, and animalistic. The unmistakable controlling image of the “ghetto chick” embodies these attitudes. The attitudes partygoers fictively deployed were precisely those that sustain the unjust treatment of actual Black people. As such, engaging in these imaginings counts as a proper “move” in accordance with a system of anti-Black oppression. In McGowan’s terms, the attitudes contemporary blackface fictively deploys “abide by” the norms of an overarching system of racial oppression. Acts of blacking up thereby bring it to bear on local situations. Thus, what practitioners of blackface do with these imaginings is enact norms of anti-Black treatment; they license people to mobilize derogatory jokes, stereotypes, and attitudes toward Black people within the bounds of that social interaction, whether or not this actually happens.
Of course, individuals who don blackface may claim, perhaps truthfully, that they intend it without serious import. The Compton Cookout’s participants provide one example. And though authorial intent does not necessarily determine what an imagining does, we may grant that many such imaginings are not rightly interpreted as endorsing the attitudes deployed. Nevertheless, by merely fictively deploying these attitudes, such uses of blackface constitute an oppressive “move” against the background of anti-Black oppression.
Licensing typically involves controlling images. Nonetheless, licensing, like normalizing described in the previous subsection, is distinct from realizing described in Section IV.A.95 Because controlling images function to justify oppressive systems, they are “ready-made” for this purpose. And because they typically circulate widely, the attitudes they fictively deploy are widely recognized; hence, they are easily wielded to bring wider oppressive systems to bear on particular microcontexts.
In sum, some imaginings constitute ethically flawed acts when used to oppress others, as moves in the game of oppression described by McGowan.96 This is so even if there is nothing ethically objectionable about the “bare prompting to fictively imagine” attitudes about some group as such, negative or otherwise.97 For it is just when these imaginings occur within a relevantly oppressive macrocontext that they enact those norms in the microcontext, thereby becoming oppressive acts.
V. How to Resist Controlling Images
The imaginatively lax, who claim that imaginings must endorse an unethical attitude to be intrinsically unethical, might protest that our view is too strong. It condemns morally permissible uses of blackface that fictively deploy their unethical attitudes precisely to undermine them. Such imaginings might be self-referential satire, or engaged in an educational setting, as when screening The Birth of a Nation in a film history class. This would appear to threaten our account, since it ought to be possible to use blackface in these ways without thereby performing acts of anti-Black oppression.
The threat, however, is illusory. Successful, critical uses of such imaginings do not reinscribe controlling images; they negate them. This is just as Collins’s vivid discussions of controlling images, themselves sometimes invitations to imagine, do not reinscribe those images. If a satirical or educational use of a potentially oppressive imagining did not negate the unethical attitudes it fictively deploys, it would simply cease to count as (successful) satire or education. Such critically framed imaginings thus differ in their content. This is much as Vermeer’s Woman with a Water Jug,98 despite depicting a map of the Netherlands hanging behind the painting’s subject, is not itself a map of the Netherlands. Such imaginings, when acts, constitute a different kind of act than uncritical imaginings fictively deploying the same attitudes. Certainly, determining just what negation requires is a substantive and difficult question. Satire alone is commonly thought impossible to define.99 Whether a particular imagining successfully negates an attitude is typically a matter for careful first-order interpretation that cannot be settled from the theoretical heights. Further confounding things, an imagining might successfully negate one controlling image while reinforcing another.100 Determining an imagining’s “illocutionary” force is difficult and requires, among other things, understanding the micro- and macrocontexts in which it occurs. Following Collins, we take the critical use of controlling images to be vital work that can be done, though it carries the possibility of misfire.101
Nevertheless, we can venture something useful from the theoretical heights to distinguish the general mechanisms for inscribing versus negating controlling images. The analogy between imagining and speech, particularly the latter’s logical form, is again helpful here.
Take some morally noxious proposition p. Asserting p is morally criticizable. One might, however, think that merely pretending to assert p, sticking a fictionality operator before p, thereby precludes any moral crime; doing so is no longer, as such, to assert p. Pursuing the analogy further, one might even point to other modal operators to pump this intuition: when we say “Donald believes that p,” we do not thereby assert p, and thus we do nothing wrong. This is the analogous position of the imaginatively lax who take mere pretence, fiction, or imagining to be morally absolving by isolating the imagined content, so to speak, from the actual world. Sticking with the analogy, our position is that sticking a fictionality operator in front of the otherwise obnoxious p does not make an utterance innocent. We might, in turn, pump a different intuition by considering other modal operators; saying “It’s possible that p” clearly does not avoid moral difficulty. One must fully deny the proposition and the sentiment its assertion expresses by negating it: “It is not the case that p.” Returning to the imagination, the problem with fictively deploying oppressive attitudes is that doing so without full-blooded negation still pumps them into the air, so to speak, polluting the atmosphere as controlling images.
Here the imaginatively strict—who claim that imaginings merely fictively deploying unethical attitudes are intrinsically unethical—might push back. If everything we have said is right, why does the moral hazard not also arise when one negates a controlling image? After all, in a parody, we still have to fictively deploy the morally fraught attitudes. Why does this not count as reinscription too? The analogy is helpful here once again. When one negates some problematic proposition p by asserting “not p,” one must still invoke p as part of what one communicates. But, ordinarily, one to no extent reinforces the sentiment behind p by doing so; on the contrary, one undermines the sentiment. Analogously, when one uses an imagining to successfully negate a controlling image, by exposing it to critical scrutiny, ridicule, or pastiche, one undermines its contribution to our shared cultural stock. In deflecting these worries, our view again strikes a middle path between imaginative strictness and laxity.
Returning to blackface, one might nonetheless wonder whether what we call “critical” blackface—that is, blackface that successfully disavows the practice and the “whitely” expectations that undergird it102—is even possible in the post–civil rights era. One might wonder this even if one accepts the middle path we have trodden. Has racism so calcified blackface that it is no longer supple enough to serve egalitarian ends? If not, then mere instantiation of blackface, however intended or executed, would deploy fictional attitudes constituting controlling images and therefore oppress. This would seem to pose a problem for us. If one could not negate the attitudes fictively deployed by blackface, this would suggest that our account of negation, including its possibility, is mistaken.
Certainly, using blackface as a means of criticizing anti-Black racism is a high-risk endeavor. For most people, we suspect, doing so would be reckless for at least two reasons. First, and perhaps most seriously, there is a high probability of misfire; it is difficult in the post–civil rights era to use blackface, even with critical intent, in a way that avoids inadvertently fictively deploying the concomitant troubling attitudes. In this, blackface is hardly unique. Compare uses of the disrobed female form to criticize objectification; often, the attempted criticism inadvertently objectifies in its own way.103 Second, even if one succeeds in this first regard by producing something that ideal interpreters of one’s imagining would deem successful criticism, one still runs a pronounced and foreseeable risk of being misunderstood. Blackface, though alive and well, has become a visual shorthand for explicit racist attitudes from a bygone era. It is difficult, therefore, to invoke its images without signaling acquiescence to those attitudes, at least to less-than-ideal interpreters. And plausibly, there is something approaching a prima facie obligation not to perform acts one fully expects will be widely misunderstood as deeply offensive.104
Nonetheless, we are optimistic about the form’s subversive potential when used judiciously. First, there appear to be successful instances of critical blackface. We cite three here.
Spike Lee’s Bamboozled is perhaps the best example. The film depicts a Black television writer, Pierre Delacroix, working for a casually racist boss in an industry unwilling to deviate from hackneyed representations of Black people. In an attempt to end his frustration, Delacroix concocts a plan to get himself fired. He devises a twenty-first-century minstrel show using Black actors in blackface. Far from getting Delacroix fired, however, the show is piloted and becomes wildly successful, forcing Delacroix and his collaborators to pitch their newfound success against their integrity. The film deftly deploys numerous artistic techniques, including literal blackface minstrelsy, to explore the horror of blackface and the social conditions that funnel Black creatives into a circumscribed set of often compromising roles. It thereby effectively satirizes both.
Another example is a comedy sketch from the television show Key & Peele. In it, two Black men insert themselves into an all-White Confederate reenactment by exaggeratedly acting like docile, enslaved caricatures. The chief reenactor, unwittingly adverting to another caricature of Black social justice warrior, interprets their act as one of antiracist protest. He defends the reenactment, refusing to continue the pretense with the enslaved characters. Finally, he relents, impatiently allowing the “slaves” into the reenactment’s fictional world before inadvertently and tellingly uttering a racial slur. At this point, the interlopers, invoking a third stereotype of Black criminality, feel forced to rob the reenactors in retaliation for their bigotry. The Confederate reenactors’ refusal to allow the Black men to participate on their own terms, the expression of hostility once they begrudgingly do allow them, and the provocative invocation of various reductive tropes all combine to bluntly satirize nostalgia for the old South. More subtly, the sketch also explores how White expectations force conformity in African Americans.
A third example, understanding blackface more broadly, is Kara Walker’s exhibition A Subtlety.105 The exhibition featured a massive foam and sugar sculpture of a hypersexualized “Mammy” lying in the pose of a sphinx, with servant children, sculpted in molasses, dotted around it. Erected in a disused sugar factory and made to look as though composed entirely of white sugar, it is recognizably a Mammy. Yet, with her breasts and vulva exposed, she horrifically commingles this long-standing matronly archetype with the licentious one of the Jezebel. The name, A Subtlety, references the sugar sculptures served by nobility at medieval banquets and the exploitative processes of sugar refinement that lay behind them, even before the transatlantic slave trade’s advent. Just as enslaved Black bodies were used in industrial sugar refinement to produce a delicacy for the privileged classes, so too have stereotypical representations of Black bodies been refined to make them, like sugar, palatable to those in power. Walker’s sculpture effectively juxtaposes these two ideas of refinement, fictively deploying the stereotypical representations as a way of rejecting them. The fact that the stereotyped representations appear to be united in an impermanent medium, sugar, in the form of a paradigmatic ruin, a sphinx, in a disused factory facing imminent demolition invites us to banish them to the past.
A second reason for retaining some defense of blackface is that we suspect that confounding variables may be driving the hard-line intuition that no blackface can serve critical ends. If someone flies a Confederate flag or spits on someone’s grave, she may thereby do any number of criticizable things. She may express or endorse a morally criticizable attitude, knowingly offend people (as just discussed), fail to respect humanity as an end, and so on. Ordinarily, however, she does not thereby deploy imaginings. The constitutive wrong of blackface we have articulated is grounded in imaginings, however, and the way these tessellate with oppressive systems. That is to say, there are potentially many wrong-making mechanisms at work. We have only argued for a small set of these: realizing, normalizing, and licensing oppression via controlling images. When someone successfully satirizes blackface and its broader presuppositions by using it, we claim, she does not commit any wrong from this set. Whether other wrongs are committed when doing so, either necessarily, typically, or frequently, is a possibility we cannot rule out. The plausibility of such other wrong-making mechanisms, however, is likely to muddy intuitions about the possibility of critical blackface, making the hard-line position tempt more than it should.
Similar remarks apply to educational cases. When an imagining fictively deploys unethical attitudes to demonstrate what not to do, this ordinarily suffices to negate them. But such educational imaginings’ illocutionary force also depends on context. Many have criticized the mass media’s reliance on “gratuitous rape” scenes—graphic, eroticized, and ubiquitous depictions of rape.106 Such depictions may serve educational ends insofar as they prescribe condemnatory attitudes toward rape. But they may also mask an underlying pornographic pleasure in female domination, especially in genres primarily targeting men. Our account illuminates how such imaginings, even when not endorsing rape, may still function “illocutionarily” to eroticize domination and promote rape culture.
In sum, blackface is oppressive in virtue of its content and what is done with it, both of which depend on the context in which it is executed. In our actual world, with its actual racist history, uncritically blacking up constitutes an oppressive imagining. But it is possible to imagine a world in which racism never existed. The crucial point we want to make here is that an imagining with identical content in a different sociohistorical context—one without racist social structures—would a fortiori not realize any part of an actual racist social system, nor normalize it, nor bring it to bear on a local context. This explains why, for all we have said, fictively deploying unethical attitudes is never objectionable in and of itself, which is where we differ from the imaginatively strict. Absent endorsement of the attitudes or relevant oppressive social relations that would make the imagining a controlling image, there is nothing intrinsically ethically wrong with imaginatively adopting unethical attitudes.
It is worth reminding ourselves here that while our discussion has focused largely on blackface, our argument is perfectly general. In addition to explaining blackface, gratuitous representations of rape, and so on, we can now also explain why the toy examples with which we began rightly elicit different intuitions. While the SAE song fictively deploys unethical attitudes that realize, normalize, and license actual oppression, the Portsmouth song deploys unethical attitudes that do not. Hence, as concerns the reasons we have outlined, only SAE’s song is ethically objectionable.
VI. Objections and Replies
Here we briefly consider some potentially lingering objections, thereby rounding out the account.
As the examples we discuss show, some imaginings are private in the strongest sense: they take place inside an individual’s head. The remainder, meanwhile, are more or less public. This distinction is ethically important, if only because of the differing scale of the potential harms such imagining types might produce. One objection is that our account fails to acknowledge it.
The distinction is morally important yet doubly irrelevant to our argument. First, potential harms concern extrinsic ethical features of imaginings, namely their causal effects, whereas our argument concerns intrinsic ones. Second, our concern is whether imaginings exhibit this intrinsic moral flaw, not to what degree they do this.
Nevertheless, one might accept our argument in spirit but still consider private imaginings exempt from ethical criticism precisely because they are private. George Sher has argued for this regarding private mental states generally,107 though others demur.108 We agree with Sher’s rejection of any moral requirement to “think only good thoughts.”109 We also agree that, in an important sense, “no thoughts or attitudes are either forbidden or required,”110 at least as concerns the imagination. This is why our account accommodates ethically permissible imaginings fictively deploying ethically unsavory attitudes, provided that they do not hook up with oppressive social structures as described. Sher himself acknowledges that private mental events can be appropriately morally criticized when they connect appropriately with public events,111 possibly even when this connection is quite weak.112 Our account articulates one way such a public-private nexus can be formed.
The imaginatively strict might wonder why it should matter whether one fictively deploys unethical attitudes in a relevantly oppressive context. Would imaginatively adopting misogynistic attitudes in a genuinely gender-egalitarian society, for instance, not be equally criticizable? If so, our diagnosis of the opening toy examples would be incorrect.
The objection misunderstands oppression, which comprises systematically patterned social structures across multiple domains of life. Rival sports teams holding negative attitudes or engaging in discriminatory behavior toward their opponents do not thereby oppress. This is because sports fans are not ordinarily disadvantaged across the multiple dimensions of health, education, employment, cultural and aesthetic representation, legal standing, and so on, on account of team preference. The same holds for contemporary derogatory imaginings of, say, ancient Phoenicians. Despite being oppressed under Babylonian rule, their subjugation has no meaningful purchase on contemporary social reality (ignoring trivial cases where “Phoenician” functions metonymically for currently oppressed groups). Imaginings fictively deploying but not endorsing would-be oppressive attitudes in relevantly nonoppressive societies, then, would not oppress for lack of enabling conditions. Isolated individuals engaging in such imaginings might fail ethically for other reasons, for example, by manifesting vicious character flaws or causing harmful consequences. Nevertheless, they will not thereby exhibit the sort of intrinsic ethical flaw we have identified. The temptation to think otherwise is, we suspect, due primarily to the fact that the most salient examples of unethical imaginings populating the nascent literature do concern oppressed groups.
Likewise, defenders of imaginative laxity might dig in. If any fictive deployment of controlling images is ethically blemished, this constraint on what one may imagine might appear “like so much fearmongering, sanctimoniousness, or prudishness.”113
Such concerns about thought policing are overcooked, however. First, there is the exception for critical uses described in Section V. Second, even in relevantly oppressive contexts, imaginings suffering the ethical flaw identified need not be absolutely morally forbidden, let alone legally prohibited. Other values—aesthetic, epistemic, and indeed other ethical considerations—might be overriding. What our discussion shows is a particular kind of pro tanto ethical reason against engaging in such imaginings. This reason does not depend on hard-to-determine causal effects, nor on attitudes the imaginings endorse. It is an important reason. But how one weighs it against competing considerations in particular cases is complicated. Moreover, our account need not condemn everyone engaging in an objectionable imagining. Apportioning blame and responsibility in oppressive contexts is difficult, especially when our psyches brim with oppressive attitudes and dispositions.114 Imaginers blacking up—even without intending to evoke or even recognizing its oppressive history—are almost always doing something oppressive with that imagining in their particular sociohistorical context. Some may have justifications or excuses, such as nonculpable ignorance, which free them from blameworthiness.
In short, our account permits criticizing imaginings fictively deploying certain attitudes in our current context. Yet it is not vulnerable to the charge of overmoralizing the imagination, since it is silent on such fictive deployments at faraway possible worlds. For all we have argued, merely imagining that a racialized group has certain stereotypical traits, that women enjoy being raped, and so on, is, in principle, ethically innocent. However, this is not the case in our sociohistorical context or others relevantly like it.
VII. Conclusion
We have argued that imaginings are intrinsically ethically flawed when they realize part of an oppressive system, normalize such a system, or license oppressive behavior in local contexts. Crucially, imaginings can do this, even when the problematic attitudes they deploy are only deployed fictively, “for fun.” This means that imaginings need not endorse such attitudes in order to be intrinsically ethically flawed. On the other hand, we have also argued that such imaginings need not be so flawed when no relevantly oppressive systems obtain. In this way, our account avoids overmoralizing the imagination.
This account yields an ethical criticism of oppressive imaginings not grounded in harmful causal effects,115 nor in endorsed attitudes. In doing so, we hope to have furnished a resource for rehabilitating our collective imaginations and realizing a future when, in Maya Angelou’s words, “the curtain falls on the minstrel show of hate; and faces sooted with scorn are scrubbed clean.”116
Notes
* This article is the equal and joint work of both authors. Special thanks to audiences at the University of Leeds’s “Race & Aesthetics” conference; the Seminario de Trabajo en Curso at the Instituto de Investigaciones Filosóficas, UNAM; the Work in Progress seminar at Yale-NUS; the XII Inter-University Workshop on Mind, Art and Morality: “Language and Power” at the University of Barcelona; the Travels in Trans-Sensoriality Symposium at LaSalle College of the Arts; the annual meetings of the British Society of Aesthetics, the European Society for Aesthetics, and the American Society for Aesthetics; Auburn University’s Aesthetics Forum; the University of Michigan’s Aesthetics Discussion Group; the University of Georgia Department of Theatre and Film Studies; and the UNED’s Philosophy Department. Thanks also to the editorial team at Ethics and especially the referees for their superlative feedback.
1. For brevity, we treat these as equivalent.
2. Bouke de Vries, “Black Pete, King Balthasar, and the New Orleans Zulus: Can Black Make-Up Traditions Ever Be Justified?,” Journal of Controversial Ideas 1 (2021): 1–14, appears to be the only other recent philosophical publication aiming at something similar. However, de Vries only sets himself the task of refuting the claim that blackface is racist (or that it is categorically wrong—de Vries appears to recognize that these two claims are not equivalent yet does not distinguish them). He laments how claims that blackface is racist are insufficiently justified. Our article might provide such justification, though this depends on how one understands racism. We take no stand here. Nor do we assert that blackface should never be performed, all things considered; see Sec. VI.
3. Our view, again, is that certain imaginings suffer a pro tanto ethical flaw, not that they are all-things-considered impermissible, nor that those who engage in them are necessarily culpable. See Sec. VI.
4. See Kendall L. Walton, Mimesis as Make-Believe (Cambridge, MA: Harvard University Press, 1990); and Berys Gaut, Art, Emotion and Ethics (Oxford: Oxford University Press 2007). For subtleties, see Nils-Hennes Stear, “Fatal Prescription,” British Journal of Aesthetics 60 (2020): 151–63, 157–59.
5. How does a work or imagining, rather than its author(s), prescribe anything? We treat this as attributable to an “implied,” “postulated,” or “manifested” author, distinct from the actual author, who (roughly) personifies the norms of proper engagement with, and interpretation of, the work. See Wayne C. Booth, A Rhetoric of Fiction (Chicago: University of Chicago Press, 1961); Alexander Nehamas, “The Postulated Author: Critical Monism as a Regulative Ideal,” Critical Inquiry 8 (1981): 133–49; and Gaut, Art, Emotion and Ethics.
6. See, respectively, Gaut, Art, Emotion and Ethics, 9; Daniel Jacobson, “In Praise of Immoral Art,” Philosophical Topics 25 (1997): 155–99, 167; Noël Carroll, “Moderate Moralism,” British Journal of Aesthetics 36 (1996): 223–38, 233; A. W. Eaton, “Where Ethics and Aesthetics Meet: Titian’s Rape of Europa,” Hypatia 18 (2003): 159–88, 171.
7. The Birth of a Nation, dir. D. W. Griffith (David W. Griffith Corp., 1915).
8. A Clockwork Orange, dir. Stanley Kubrick (Polaris Production and Hawk Films, 1971).
9. Glenn Carrick and David Batty, “In Abu Dhabi They Call It Happiness Island. But for the Migrant Workers, It Is a Place of Misery,” Guardian, December 22, 2013, https://www.theguardian.com/world/2013/dec/22/abu-dhabi-happiness-island-misery/.
10. See Berys Gaut, “Art and Ethics,” in The Routledge Companion to Aesthetics, ed. Berys Gaut and Dominic Lopes (London: Routledge, 2013), 394–403, 395; Jacobson, “In Praise of Immoral Art,” 165; Eaton, “Where Ethics and Aesthetics Meet,” 174–75; James Harold, “On Judging the Moral Value of Narrative Artworks,” Journal of Aesthetics and Art Criticism 64 (2006): 259–70, 260; Adriana Clavel-Vázquez, “Rethinking Autonomism: Beauty in a World of Moral Anarchy,” Philosophy Compass 13 (2018): 1–10, 2. For a dissenting view, see Ted Nannicelli, Artistic Creation and Ethical Criticism (New York: Oxford University Press, 2020).
11. Compare Walton’s use of ‘fiction’ in Walton, Mimesis as Make-Believe.
12. David Lewis, “Extrinsic Properties,” Philosophical Studies 44 (1983): 197–200.
13. Artemisia Gentileschi, Judith Slaying Holofernes, 1612–13, oil on canvas, Museo di Capodimonte, Naples.
14. Even this is dubious: w’s representing x may require x’s existence, w’s similarity to x and/or an authorial intention to represent x in w, etc., thereby counting only as weakly intrinsic. Indeed, the vast majority of interesting intrinsic features of imaginings, like artworks, are only weakly intrinsic, especially if aesthetic properties are response dependent, as many maintain.
15. See James Harold, “Flexing the Imagination,” Journal of Aesthetics and Art Criticism 61 (2003): 247–57; Susan Feagin, “Film Appreciation and Moral Insensitivity,” Midwest Studies in Philosophy 34 (2010): 20–33, 31–32; A. W. Eaton, “Robust Immoralism,” Journal of Aesthetics and Art Criticism 70 (2012): 281–92; Aaron Smuts, “The Ethics of Singing Along: The Case of ‘Mind of a Lunatic,’” Journal of Aesthetics and Art Criticism 71 (2013): 121–29. For arguable further examples, see also G. E. Moore, Principia Ethica (1903; repr., Cambridge: Cambridge University Press, 1922), 208–10; and Talbot Brewer, The Bounds of Choice: Unchosen Virtues, Unchosen Commitments (New York: Garland, 2000), 38–39n. Also among this group are Gaut, Art, Emotion and Ethics; and Alan Hazlett, “How to Defend Response Moralism,” British Journal of Aesthetics 49 (2009): 241–55, if actual fiction-directed emotions collapse into fictional or imagined emotions, as they would on Kendall Walton’s influential theory in Mimesis as Make-Believe.
16. ‘Endorsement’ is, on the one hand, such a straightforward term that the British Board of Film Classification uses it in the relevant sense in its guidelines without explanation. See British Board of Film Classification, BBFC Classification Guidelines (London: British Board of Film Classification, 2009). On the other hand, it is potentially ambiguous. See Justin D’Arms and Daniel Jacobson, “Sentiment and Value,” Ethics 110 (2000): 722–48. Our view applies for all standard disambiguations.
17. See James Harold, “Immoralism and the Valence Constraint,” British Journal of Aesthetics 48 (2008): 45–64, 51–53; Brandon Cooke, “Ethics and Fictive Imagining,” Journal of Aesthetics and Art Criticism 72 (2014): 317–27; and George Sher, “A Wild West of the Mind,” Australasian Journal of Philosophy 97 (2019): 483–96. Christopher Bartel and Anna Cremaldi, in criticizing Cooke, extend the view to fictive imaginings that function to cultivate immoral desires rather than just beliefs; see Christopher Bartel and Anna Cremaldi, “‘It’s Just a Story’: Pornography, Desire, and the Ethics of Fictive Imagining,” British Journal of Aesthetics 58 (2018): 37–50. Nils-Hennes Stear gives another argument in the same spirit; see Nils-Hennes Stear, “Sadomasochism as Make-Believe,” Hypatia 24 (2009): 21–38. Alessandro Giovanelli thinks that “a case can be made” for the view; see Alessandro Giovannelli, “Ethical Criticism in Perspective: A Defense of Radical Moralism,” Journal of Aesthetics and Art Criticism 71 (2013): 335–48, 339.
18. For examples of such ambiguous use, see Noël Carroll, “Art and Ethical Criticism: An Overview of Recent Directions of Research,” Ethics 110 (2000): 350–87; Gaut Art, Emotion and Ethics, 192–94; and Matthew Kieran, Revealing Art (New York: Routledge, 2005), 105–7.
19. Tamar Szabo Gendler, “The Puzzle of Imaginative Resistance,” Journal of Philosophy 97 (2000): 55–81.
20. US Department of State, “Memorandum of Conversation between President Ford, President Suharto, and Secretary Kissinger,” NSA Archive (1975), http://nsarchive.gwu.edu/NSAEBB/NSAEBB62/doc4.pdf; Commission for Reception, Truth, and Reconciliation, Timor-Leste (CAVR), The Report of the Commission for Reception, Truth, and Reconciliation: Executive Summary (2005), https://www.etan.org/etanpdf/2006/CAVR/Chega!-Report-Executive-Summary.pdf.
21. Cooke, “Ethics and Fictive Imagining,” 318–19.
22. See ibid. on “fictive” and “non-fictive” imagining. Roughly the same ambiguity is also identified in Gendler, “Puzzle of Imaginative Resistance,” 76–77; Gaut, Art, Emotion and Ethics, 250–51; and Stear, “Fatal Prescription.”
23. Lindy West, “A Complete Guide to ‘Hipster Racism,’” Jezebel, April 26, 2012, http://jezebel.com/5905291/a-complete-guide-to-hipster-racism/.
24. There are strong reasons not to capitalize ‘White’, such as its long use by White supremacist groups. See, e.g., Brittany Wong, “Here’s Why It’s a Big Deal to Capitalize the Word ‘Black,’” September 3, 2020, https://www.huffpost.com/entry/why-capitalize-word-black_l_5f342ca1c5b6960c066faea5/; and Nancy Coleman, “Why We’re Capitalizing Black,” New York Times, July 5, 2020, https://www.nytimes.com/2020/07/05/insider/capitalized-black.html/. With some reservations, we opt to follow the convention of capitalizing both ‘Black’ and ‘White’ as advocated by Eve L. Ewing and Kwame Anthony Appiah; see Eve L. Ewing, “I’m a Black Scholar Who Studies Race. Here’s Why I Capitalize ‘White,’” Zora, July 2, 2020, https://zora.medium.com/im-a-black-scholar-who-studies-race-here-s-why-i-capitalize-white-f94883aa2dd3/; and Kwame Anthony Appiah, “The Case for Capitalizing the B in Black,” Atlantic, June 18, 2020, https://www.theatlantic.com/ideas/archive/2020/06/time-to-capitalize-blackand-white/613159/.
25. Scott Jaschik, “Oregon: Professor in Blackface Violated Anti-harassment Policy,” Inside Higher Ed, January 3, 2017, https://www.insidehighered.com/news/2017/01/03/university-oregon-finds-professor-who-wore-blackface-party-violated-anti-harassment/.
26. See C. Richard King and David J. Leonard, “The Rise of the Ghetto-Fabulous Party,” Colorlines, October 3, 2007, https://www.colorlines.com/articles/rise-ghetto-fabulous-party/; S. E. Smith, “Hipster Racism,” This Ain’t Livin’, July 16, 2009, http://meloukhia.net/2009/07/hipster_racism/; West, “Complete Guide to ‘Hipster Racism’”; Tanya Rodriguez, “Numbing the Heart: Racist Jokes and the Aesthetic Affect,” Contemporary Aesthetics 12 (2014): 1–16; Jaschik, “Oregon: Professor in Blackface.”
27. J. L. Austin, How to Do Things with Words (Cambridge, MA: Harvard University Press, 1962).
28. In such cases, the illocutionary act succeeds while the perlocutionary act does not. Some have argued that hearer uptake—understood as a kind of minimal perlocutionary effect—is required for all successful illocutionary acts, but this is contested. See William Alston, Illocutionary Acts and Sentence Meaning (Ithaca, NY: Cornell University Press, 2000); Mark Sluys, “Getting the Message and Grasping It: The Give-and-Take of Discourse,” Philosophia 47 (2019): 207–24; and Mary Kate McGowan, Just Words: On Speech and Hidden Harm (Oxford: Oxford University Press, 2019).
29. See Catharine MacKinnon, Only Words (Cambridge, MA: Harvard University Press, 1993); Rae Langton, “Speech Acts and Unspeakable Acts,” Philosophy and Public Affairs 22 (1993): 293–330, 305–30; and Jennifer Hornsby and Rae Langton, “Free Speech and Illocution,” Legal Theory 4 (1998): 21–37.
30. Luvell Anderson, Sally Haslanger, and Rae Langton, “Language and Race,” in The Routledge Companion to Philosophy of Language, ed. Gillian Russell and Delia Graff Fara (New York: Routledge, 2012), 753–67.
31. Jennifer Saul, “Pornography, Speech Acts and Context,” Proceedings of the Aristotelian Society 106 (2006): 229–48.
32. Many philosophers have invoked speech act theory to explain the nature of fiction. See, e.g., John Searle, “The Logical Status of Fictional Discourse,” New Literary History 6 (1975): 319–32; Gregory Currie, “Works of Fiction and Illocutionary Acts,” Philosophy and Literature 10 (1986): 304–8; Gregory Currie, The Nature of Fiction (Cambridge: Cambridge University Press, 1990); and Aloysius Martinich, “A Theory of Fiction,” Philosophy and Literature 25 (2001): 96–112. There are serious problems with these approaches to understanding fiction: (1) many works of fiction, e.g., painting and program music, do not involve speech or even presuppose the possibility of language (see Walton, Mimesis as Make-Believe, 75–89); (2) imaginings do not satisfy the first two conditions that utterances must always involve the phonetic act of uttering certain noises and the phatic act of uttering those noises that follow a particular vocabulary and grammar on Austin’s definition of what it is “to say anything” (Austin, How to Do Things, 92–93); (3) philosophy of language orthodoxy treats speech as fundamentally communicative and speaker intentions as essential to this (e.g., Austin, How to Do Things; P. F. Strawson, “Intention and Convention in Speech Acts,” Philosophical Review 73 [1964]: 439–60; and John Searle, Speech Acts: An Essay in the Philosophy of Language [Cambridge: Cambridge University Press, 1969]), whereas it is more widely accepted in aesthetics that intentions are often important but rarely overriding in determining fictional content (again, see, e.g., Walton, Mimesis as Make-Believe, 75–89). See Kathleen Stock, Only Imagine (Oxford: Oxford University Press, 2017), for criticism of the orthodoxy.
33. For more on how fictions assert, see Lewis, “Extrinsic Properties”; Rae Langton and Caroline West, “Scorekeeping in a Pornographic Language Game,” Australasian Journal of Philosophy 77 (1999): 303–19; and Eva-Maria Konrad, “Signposts of Factuality: On Genuine Assertions in Fictional Literature,” in Art and Belief, ed. Ema Sullivan-Bissett, Helen Bradley, and Paul Noordhof (Oxford: Oxford University Press, 2017), 42–62.
34. George Orwell, Animal Farm: A Fairy Story (London: Secker & Warburg, 1945).
35. Just as some utterances (e.g., issuing from Tourette syndrome) do not constitute speech acts, not all imaginings constitute imaginative acts; they may arise wholly unbidden or nonconsciously, disqualifying them as “prima facie candidates for ethical evaluation” (see Cooke, “Ethics and Fictive Imagining,” 318). Hence, those parts of our theory concerned with actions are restricted to deliberately undertaken imaginings unquestionably amenable to agential evaluation. Nevertheless, what we argue in Sec. IV.A makes clear why certain nonvoluntary imaginings are also ethically undesirable.
36. See, e.g., Marilyn Frye, The Politics of Reality: Essays in Feminist Theory (Trumansburg, NY: Crossing, 1983); Iris Marion Young, Justice and the Politics of Difference (Princeton, NJ: Princeton University Press, 1990); Patricia Hill Collins, Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment, 2nd ed. (New York: Routledge, 2000); and McGowan, Just Words.
37. One could, in principle, normalize or license oppressive behavior using a variety of fictional contents, some of which might be idiosyncratic to a particular situation. However, here we focus on fictional content partially constituting oppressive systems (see Sec. IV.A) and for that reason automatically lending itself to such uses.
38. Strictly speaking, the speech acts with which we are drawing an analogy here are not illocutionary acts (because they do not depend on the intention to communicate), but instead what McGowan calls “parallel acts” generated alongside illocution (see McGowan, Just Words). To minimize confusion, however, we will continue to appeal to the “illocutionary” for the purposes of our analogy, which has its limits anyway (see n. 28).
39. The constitution/causation distinction is widely recognized, if contentious. For our purposes, it is enough to distinguish them by pointing out that constitution is a paradigmatically asymmetric, synchronous relation that occurs between nonindependently existing entities, while causation is a diachronic process that occurs across time, typically between independently existing entities. To illustrate with an example from Petri Ylikoski, the questions “How did the glass become fragile?” and “Why did the glass break?” solicit causal explanations. Meanwhile, the question “What makes the glass fragile?” solicits a constitutive explanation. See Petri Ylikoski, “Causal and Constitutive Explanation Compared,” Erkenntnis 78 (2013): 277–97. For complications, see Gideon Rosen, “Ground by Law,” Philosophical Issues 27 (2017): 279–301, 280–81. Note that we are not, for our purposes, using McGowan’s concept of “harm constitution,” which refers to a specific way of causing harm (see McGowan, Just Words).
40. See, e.g., Joel Feinberg, The Moral Limits of the Criminal Law, vol. 4, Harmless Wrongdoing (New York: Oxford University Press, 1988); Arthur Ripstein, “Beyond the Harm Principle,” Philosophy and Public Affairs 34 (2006): 215–45; and Adam Slavny and Tom Parr, “Harmless Discrimination,” Legal Theory 21 (2015): 100–114.
41. Joel Feinberg, The Moral Limits of the Criminal Law, vol. 1, Harm to Others (New York: Oxford University Press, 1984), 36.
42. Catherine Cole, “American Ghetto Parties and Ghanaian Concert Parties,” in Burnt Cork: Traditions and Legacies of Blackface Minstrelsy, ed. Stephen Johnson (Amherst: University of Massachusetts Press, 2012), 18–50.
43. See Putney Swope, dir. Robert Downey Sr. (Herald Productions, 1969); and Amos ’n’ Andy, dir. Charles Barton (CBS, 1951–53). Whether other practices, such as “digital blackface,” the comic use of representations of Blackness in internet memes, or characters like Annie, the Mammy-like spokesperson for fast-food chain Popeyes, fall under this concept is less clear. See Aaron Nyerges, “Explainer: What Is Digital Blackface,” United States Studies Centre, University of Sydney, August 23, 2018, https://www.ussc.edu.au/analysis/what-is-digital-blackface/.
44. See William T. Leonard, Masquerade in Black (Metuchen, NJ: Scarecrow, 1986).
45. Nicholas Sammond, Birth of an Industry: Blackface Minstrelsy and the Rise of American Animation (Durham, NC: Duke University Press, 2015), 6, 221.
46. William T. Lhamon, Raising Cain: Blackface Performance from Jim Crow to Hip Hop (Cambridge, MA: Harvard University Press, 1998).
47. George Lippard, Blanche of Brandywine; or, September the Eleventh, 1777 (Philadelphia: Zieber, 1846).
48. Sammond, Birth of an Industry.
49. Benjamin Miller, “Twisting the Dandy: The Transformation of the Blackface Dandy in Early American Theatre,” Journal of American Drama and Theatre 27 (2015): 1–21.
50. The Jazz Singer, dir. Alan Crosland (Warner Bros. Pictures and the Vitaphone Corporation, 1927).
51. Charles Musser, “Why Did Negroes Love Al Jolson and The Jazz Singer? Melodrama, Blackface and Cosmopolitan Theatrical Culture,” Film History 23 (2011): 196–222.
52. See Justin Tosi and Brandon Warmke, “Moral Grandstanding,” Philosophy and Public Affairs 44 (2016): 197–217.
53. Frederick Douglass, The Life and Writings of Frederick Douglass, ed. Philip S. Foner (New York: International, 1950), 2:387. See also Sammond, Birth of an Industry, 1–2, 151.
54. Ralph Ellison, “Change the Joke and Slip the Yoke,” Partisan Review 25 (1958): 212–21.
55. Musser, “Why Did Negroes,” 206.
56. Tavia Nyong’o, The Amalgamation Waltz: Race, Performance, and the Ruses of Memory (Minneapolis: University of Minnesota Press, 2009), 121. See also discussion of Spike Lee’s Bamboozled in Sec. V.
57. See Ellison, “Change the Joke”; and Leroi Jones [Amiri Baraka], Blues People: Negro Music in White America (1963; repr., New York: HarperCollins, 1999).
58. Collins, Black Feminist Thought.
59. Ibid., 69.
60. Much has been written on “ideology” and the “social imaginary.” Both literatures capture similar ideas. See, e.g., Tommie Shelby, “Ideology, Racism, and Critical Social Theory,” Philosophical Forum 34 (2003): 153–88; and Cornelius Castoriadis, The Imaginary Institution of Society (1975; repr., Cambridge, MA: MIT Press, 1998). Blackface’s relation to a “racial imaginary” is discussed in Sammond, Birth of an Industry, as elsewhere.
61. Controlling images are parts of what Sewell calls “schemas,” which pair with “resources” to make up social structures. See William H. Sewell Jr., “A Theory of Structure: Duality, Agency, and Transformation,” American Journal of Sociology 98 (1992): 1–29. On Haslanger’s practice-first account, controlling images are some of the social meanings that make up a cultural technē. See Sally Haslanger, “I—Culture and Critique,” Supplement to the Proceedings of the Aristotelian Society 91 (2017): 149–73.
62. Eduardo Bonilla-Silva, “The Structure of Racism in Color-Blind, ‘Post-racial’ America,” American Behavioural Scientist 59 (2015): 1358–76, 1361.
63. Collins, Black Feminist Thought. See also Antonio Gramsci, Selections from the Prison Notebooks of Antonio Gramsci, ed. and trans. Q. Hoare and G. Nowell Smith (New York: International, 1971).
64. See Young, Justice, who identifies “cultural imperialism” as one of five irreducible faces of oppression. Young suggests that experiencing any one face—exploitation, marginalization, powerlessness, and violence being the other four—is sufficient for a group to count as oppressed, but she also notes that “nearly all, if not all, groups … said to be oppressed suffer cultural imperialism” while varying on the others (ibid., 64). This suggests that cultural imperialism, enacted in the hegemonic domain, plays a special role in sustaining each of the other four faces of oppression. Moreover, the other four faces each operate across all three of Collins’s other domains of power. Hence, we think that Young’s alternative framework for understanding oppression is compatible with our argument.
65. Frye, Politics of Reality. See also Young, Justice; and Sally Haslanger, “Oppressions: Racial and Other,” in Racism in Mind, ed. Michael P. Levine and Tamas Pataki (Ithaca, NY: Cornell University Press, 2004), 97–123.
66. These are “(1) accounting for the existence of racial inequality; (2) providing basic rules on engagement in interracial interactions; (3) furnishing the basis for actors’ racial subjectivity; (4) shaping and influencing the views of dominated actors; and (5) by claiming universality, hiding the fact of racial domination.” What we here call “normalizing” corresponds to (1), (3), and (5), while “licensing” corresponds to (2). See Bonilla-Silva, “Structure of Racism,” 1361.
67. To be sure, there may be causal processes involved in enacting a fact: moving one’s vocal chords causes air particles to move, etc. But this is not the mechanism enacting the fact that one said something.
68. McGowan, Just Words.
69. Ibid., 21.
70. For an example of how a positive stereotype, i.e., being sexually attractive, normalizes racially fetishized groups, see Robin Zheng, “Why Yellow Fever Isn’t Flattering: A Case against Racial Fetishes,” Journal of the American Philosophical Association 2 (2016): 400–419.
71. Blair L. M. Kelley, “A Brief History of Blackface,” Grio, October 30, 2013, http://thegrio.com/2013/10/30/a-brief-history-of-blackface-just-in-time-for-halloween/.
72. Douglas A. Jones, The Captive Stage: Performance and the Proslavery Imagination of the Antebellum North (Ann Arbor: University of Michigan Press, 2014), 9; see also Nyong’o, Amalgamation Waltz, 122.
73. Howard Zinn, A People’s History of the United States: 1492–Present (1980; repr., New York: Routledge, 2015), 194.
74. William Shakespeare, The Tragedy of Othello, the Moor of Venice (London: Globe Theatre, 1601–4).
75. As one theater director stated concerning a production of the Verdi opera based on Shakespeare’s play, “It really did seem very obvious given our cultural history and political history in the United States, that … the idea of putting [Otello] in blackface was completely unthinkable.” See Nicky Woolf, “Decision to Scrap Blackface from Otello Not Complicated, Says Met Director,” Guardian, September 22, 2015, https://www.theguardian.com/music/2015/sep/22/otello-metropolitan-opera-scraps-blackface/.
76. Some morris dancing troupes have recently substituted different colors for traditional black face paint. See PA Media, “May Day Morris Dancers Swap Black Face Paint for Blue over Concerns of Racism,” Guardian, May 1, 2021, https://www.theguardian.com/world/2021/may/01/may-day-morris-dancers-swap-black-face-paint-for-blue-over-concerns-of-racism/. The “morris” in “morris dance” almost certainly derives from “moorish.” Nonetheless, the connection to race is more complicated than the etymology implies. See John Forrest, The History of Morris Dancing, 1458–1750 (Toronto: University of Toronto Press, 1999), 3–26, 363–67.
77. Melissa Noel, “The Fight against ‘Black Pete,’ a Holiday Tradition,” NBC News, December 23, 2015, http://www.nbcnews.com/news/nbcblk/fight-against-black-pete-holiday-blackface-tradition-n485081/; Camilo Smith, “A Latino Tradition for Three Kings Day Stirs Controversy,” Houston Chronicle Blog, January 4, 2016, http://blog.chron.com/lavoz/2016/01/a-latino-tradition-for-three-kings-day-stirs-controversy/.
78. See Mary Kate McGowan, “Conversational Exercitives and the Force of Pornography,” Philosophy and Public Affairs 31 (2003): 155–89; Mary Kate McGowan, “On Locker Room Talk and Linguistic Oppression,” Philosophical Topics 46 (2018): 165–81; and McGowan, Just Words.
79. Robert Mark Simpson, “Un-ringing the Bell: McGowan on Oppressive Speech and the Asymmetric Pliability of Conversations,” Australasian Journal of Philosophy 91 (2013): 555–75, 563.
80. McGowan, “Conversational Exercitives”; McGowan, “On Locker Room Talk”; McGowan, Just Words.
81. McGowan, Just Words, 399.
82. McGowan demurs here. For her, constituting harm by enacting oppressive norms is itself a specific way of causing harm: harm results from behavior conforming to the norm enacted. See McGowan, “On Locker Room Talk”; and McGowan, Just Words. We agree, of course, that imaginings can cause oppression—that they can “perlocutionarily” harm in this way. But we wish to show that they can oppress even absent this. This is an ongoing point of debate in the literature: a third view is that enacting oppressive norms is itself a harm irrespective of downstream consequences. See, e.g., Katharine Jenkins, “Ontic Injustice,” Journal of the American Philosophical Association 6 (2020): 188–205, who contends that it is a kind of harmful “moral injury” that damages an individual’s dignity.
83. Indeed, McGowan explicitly takes the mechanism she identifies to apply to jokes, storytelling, and other imaginings, which she calls “nonserious speech” (see McGowan, “Conversational Exercitives,” 183–85).
84. Paul Taylor, Black Is Beautiful: A Philosophy of Black Aesthetics (Chichester: Wiley, 2016), 182.
85. Bamboozled, dir. Spike Lee (40 Acres and a Mule Filmworks, 2000).
86. Nathan Irvin Huggins, Harlem Renaissance (1972; repr., New York: Oxford University, 2007), 258, 274–75, 279–86.
87. Scott Herring, “Du Bois and the Minstrels,” MELUS 22 (1997): 3–17, 10. This is an example of what James Scott terms a “weapon of the weak.” See James C. Scott, Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven, CT: Yale University Press, 2008).
88. Huggins, Harlem Renaissance, 260.
89. William F. Stowe and David Grimsted, “White-Black Humor,” Journal of Ethnic Studies 3 (1975): 78–96; Huggins, Harlem Renaissance, 267–74; Stephen Johnson, introduction to Burnt Cork: Traditions and Legacies of Blackface Minstrelsy, ed. Stephen Johnson (Amherst: University of Massachusetts Press, 2012), 7.
90. Sammond, Birth of an Industry, 6.
91. Huggins, Harlem Renaissance; Stowe and Grimsted, “White-Black Humor”; Dale Cockrell, Demons of Disorder: Early Blackface Minstrels and Their World (New York: Cambridge University Press, 1997); Musser, “Why Did Negroes”; William T. Lhamon, “Turning around Jim Crow,” in Burnt Cork: Traditions and Legacies of Blackface Minstrelsy, ed. Stephen Johnson (Amherst: University of Massachusetts Press, 2012), 18–50.
92. Eric Lott, Love and Theft: Blackface Minstrelsy and the American Working Class (New York: Oxford University Press, 2013).
93. Sammond, Birth of an Industry, 262. See also Jack Limon, Stand-Up Comedy in Theory; or, Abjection in America (Durham, NC: Duke University Press, 2000); and Adrienne Rich, “Taking Women Students Seriously,” Radical Teacher 11 (1979): 40–43, 41.
94. Lisa Wade, “The Compton Cookout: Racism, Resistance, and Backlash,” Sociological Images, February 28, 2010, https://thesocietypages.org/socimages/2010/02/28/the-compton-cookout-racism-resistance-and-backlash/.
95. See n. 37.
96. See Taylor, Black Is Beautiful, 58–62.
97. Cooke, “Ethics and Fictive Imagining,” 322.
98. Johannes Vermeer, Woman Holding a Water Jug, 1660–62, oil on canvas, Metropolitan Museum of Art, New York.
99. Dieter Declercq, “A Definition of Satire (and Why a Definition Matters),” Journal of Aesthetics and Art Criticism 76 (2018): 319–30, 325.
100. See bell hooks, Black Looks: Race and Representation (Boston: South End, 1992), 117–18.
101. Collins, Black Feminist Thought, 27, 77, 86–88, 101, 142–43, 156–57, 169, 173–76.
102. Marilyn Frye, “White Woman Feminist,” in Moral Issues in Global Perspective, ed. Christine Koggel (Peterborough: Broadview, 1992), 2:116–28.
103. See Feagin, “Film Appreciation,” 25–26, for an example from a Robert Altman film.
104. These two possibilities appear to map onto Luvell Anderson’s last two categories of jokes concerning race: “racially insensitive” and “racist” jokes. See Luvell Anderson, “Why So Serious? An Inquiry on Racist Jokes,” Journal of Social Philosophy 1 (2020): 1–16, 9–10.
105. Kara Walker, A Subtlety: An Homage to the Unpaid and Overworked Artisans Who Have Refined Our Sweet Tastes from the Cane Fields to the Kitchens of the New World on the Occasion of the Demolition of the Domino Sugar Refining Plant, 2014, Domino Sugar Refinery, New York City.
106. Maureen Ryan, “The Progress and Pitfalls of Television’s Treatment of Rape,” Variety, December 6, 2016, http://variety.com/2016/tv/features/rape-tv-television-sweet-vicious-jessica-jones-game-of-thrones-1201934910/.
107. Sher, “Wild West of the Mind.”
108. See Robert Merrihew Adams, “Involuntary Sins,” Philosophical Review 94 (1985): 3–31; and Rima Basu, “What We Epistemically Owe to Each Other,” Philosophical Studies 176 (2019): 915–31.
109. Sher, “Wild West of the Mind,” 490.
110. Ibid., 484.
111. Ibid., 485.
112. Ibid., 494.
113. Cooke, “Ethics and Fictive Imagining,” 325.
114. Recent theorists have argued that structural oppression requires no oppressors, or that agents can and do participate in oppression without individual culpability. See Haslanger, “Oppressions”; Iris Marion Young, Responsibility for Justice (New York: Oxford University Press, 2011); Robin Zheng, “Bias, Structure, and Injustice: A Reply to Haslanger,” Feminist Philosophy Quarterly 4 (2018): 1–30.
115. On the desirability of such tools, see A. W. Eaton, “A Sensible Anti-porn Feminism,” Ethics 117 (2007): 674–715.
116. Maya Angelou, “A Brave and Startling Truth,” in Maya Angelou, ed. Edwin Graves Wilson (New York: Sterling, 2007), 43–46.
